Computer Science
BESSTIE: A Benchmark for Sentiment and Sarcasm Classification for Varieties of English
D. Srirag, A. Joshi, et al.
BESSTIE introduces the first labelled benchmark for sentiment and sarcasm across three English varieties (en-AU, en-IN, en-UK), built from Google Places reviews and Reddit comments with manual and automatic validation. Nine large language models were fine-tuned and evaluated, revealing consistent advantages on inner-circle varieties (en-AU, en-UK) and challenges in cross-variety generalisation—especially for sarcasm. Research conducted by Dipankar Srirag, Aditya Joshi, Jordan Painter, and Diptesh Kanojia. Dataset available on Hugging Face.
Related Publications
Explore these studies to deepen your understanding of the subject.
